The version of RAW used by my camera is NEF, the Nikon raw.   There is an option to compress the data and Nikon says there will be little loss of quality.

I have used both compressed and full file and have not seen loss of quality, or at least not a loss I have recognized.       Compression is slightly less than 1/2, such as from 15mb to about 8mb.

Any thoughts on compressing RAW in camera or leaving it at full size file?        TomDart.

No idea  but I would like to see other peoples opinions

my Canon 40d has the option to shoot either RAW or sRAW, which as I understand it outputs a raw file of approx 6 megapixels, with all the raw flexibility of the other one but less of the resolution, hence smaller file size.

could be it's different with your camera/ other brands, but the smaller one should be essentially just as high in quality, just fewer megapixels to play with, so probably not the best choice if you were considering larger prints from the shot.

Adam

Based on 2 articles I found, there is some loss of image information. 

One of those articles states,

there is some loss of data, mostly in the form of lowered resolution
in the highlights.

The loss of information results from translating the output from the image sensor from 12 bits per sensor (values 0 to 4095) into non-linear coded values 0-682. 

For the details, please read,

Is the Nikon D70 NEF (RAW) format truly lossless?

Nikon Compressed Raw Format - Lossy or Lossless?

Meowgli said,

but the smaller one should be essentially just as high in quality,
just fewer megapixels to play with

Sorry but that's not true ... that is not how Nikon has engineered compressed RAW. 

The Nikon compressed RAW does not change the number of pixels. 

The Nikon RAW reduces the highlight detail (range of light/dark) in all of the pixels. 

The decision to use compressed RAW should depend on how much time is needed to record onto the flash memory.  As a test, try shooting the same scene in both RAW and compressed RAW and time the image save (you know, One thousand one; one thousand two; ... to count the seconds). 

You also might be able to get a faster flash memory. 

The type of scene will also influence the choice ... for sports and action shots, you may need to use compressed so that you won't miss a shot; for landscapes, you wouldn't use compressed ... except for pictures of violent storms [sly grin]. 

--Martin

ok, apologies for my misinformation in that case.. it seems Nikon and Canon have gone about the 'smaller raw' process a slightly different way.... I can only offer my input on what I know and much as I like Nikon stuff I confess I'm not well read up on it.... on a side note, the sRAW format seems to be a bit of a bugger in terms of compatability, not opening on certain machines etc, and from my point of view at the moment it's a bit of a waste of time as Adobe's DNG converter doesn't seem to recognise the format (and my CS2 setup doesn't recognise the 40d's RAW files straight out of camera...)

Adam

I think the phrase little loss of quality means there will be some degradation... For the web it is of little significance, but in printing this may not be the case as pointed out by Adam as there will be a reduction in detail...  While smaller file size will write to the cards faster to the camera's memory buffer Martin,  this does not mean the camera will be able to capture more shots by speeding up the frame rate at least not in Canon cameras, an option I would like to have seen in the new 5DMK11.. Withe 21 million pix it would have been nice to have say a compressed sRaw file at 6 pr 7 fps .. If Nikon have cracked that then that may be good news ..

I imagine side by side tests of compressed and uncompressed RAW would have to be made, with images of various ranges of light and shadow,  including print as a primary test medium.

You see, large file sizes are nice to have in working an image for print.  Yet, in print is the difference in "highlight detail" supposedly compressed by Nikon visible to us?   I can believe in the truth of some detail degredation but is this a degredation anyone will see?   I don't have an answer for that one.

I can shoot compressed NEF (NIkon RAW) as fast as the camera will shoot with no problem.
I do have a high speed memory cards but am not sure how much that helps since I have always purchased high speed cards...newer ones simply faster than the older fast ones.

Let's bring some thoughts together ...

Compression is slightly less than 1/2, such as from 15mb to about 8mb.

You also might be able to get a faster flash memory. 

While smaller file size will write to the cards faster to the camera's
memory buffer Martin,  this does not mean the camera will be able
to capture more shots by speeding up the frame rate

All other things being equal (unless I'm missing something), smaller file size
must mean faster possible frame rate. 

You see, large file sizes are nice to have in working an image for print. 

Yes ... ... with a jpg file, larger file size means either more pixels or less compression of color gamut.  While not the same (compression) methodology, Nikon RAW vs. compressed RAW also offers a tradeoff between file size and image detail. 

Yet, in print is the difference in "highlight detail" supposedly compressed
by Nikon visible to us?

Probably not ... EXCEPT if your post processing needs/uses that 'lost' detail. 

For example, after shooting, you find some interesting detail in an area of the image ... you crop to select that part,  adjust the levels and contrast ... and find that you don't have the shadow detail you expected.  As an example, a shot of a bird in it's nest might be in itself a great shot ... but then back at home, you see that you also have a chick emerging from the shell ... and want to make that portion of the image into a print ...

As I said before, if the scene has no motion (landscape, architecture, macro, sunset, clouds, ... ), you should shoot uncompressed. 

OTOH if the scene has action, movement, ... anything that needs a faster frame rate to get the perfect shot, decisive moment, the right expression, the foot sinking into the kicked ball, ... then you would want to shoot compressed. 

That having been said, too bad that you won't always know in advance ...

--Martin

quote  While smaller file size will write to the cards faster to the camera's

memory buffer Martin,  this does not mean the camera will be able
to capture more shots by speeding up the frame rate 

 
Martin says

"All other things being equal (unless I'm missing something), smaller file size
must mean faster possible frame rate."

The specs given by camera makers are the frame rates in Jpeg which are compressed I hardly see how a sRaw file will be quicker than the max frame rate quoted by the makers! 

Agreed, the max frame rate would be associated with a jpg file. 

OTOH, comparing RAW with compressed RAW, I would expect the frame rate for RAW to be less than the frame rate for compressed RAW. 

--Martin

Yes Martin this is what I cant undersand I would be willing to accept lower resolutions files in exchange for faster frame rates but this does not appear to be the case despite having invested in the fastest cards on the market the only benefit I can determine is the memory buffer gets fuller quicker, and uploading time to my PC

Your website lists Canon EOS 5D and Canon EOS I0D (I think you mean Canon EOS 10D ... 'one zero' not 'i zero'). 

As regards frame rate, are you talking about the Canon EOS 5D or another camera?

As regards flash memory, are you talking about the Extreme lll 2GB SanDisk 133x? 

--Martin

Yes just a typo I have both 5D and 10D and access to 1Dmk11 which I carried out tests on the recent cards are the Extreme 4 versions of the Sandisk  4GB and 8GB bought from my dealer not fake ones which under perform

Quote - Agreed, the max frame rate would be associated with a jpg file. 

OTOH, comparing RAW with compressed RAW, I would expect the frame rate for RAW to be less than the frame rate for compressed RAW. 

--Martin

For most current DSLR cameras  the max frame rate is almost the same whether shooting in  RAW or JPEG (at least if judging from the manufacturers specs).

What is usually limited by shooting RAW is the number of shots you can take before the frame rate slows down due to buffer in the camera filling up.

For instance my Sony A200 will shoot at around 3fps in both RAW and JPEG, but wheras it can shoot at this rate until the memory card is full in JPEG mode it will only take about 7 pictures before slowing down when shooting RAW even with the fastest memory cards available.

A compressed RAW format should allow you to keep shooting at maximum speed for longer than with uncompressed RAW but in most DSLRcameras will have a minimal effect on the maximum frame rate which can be achieved.

Andy

Yep thats how I  have found in practice on Canon Cameras too

My experience gained today verifies the above, partially.  I did not shoot enough to slow things down with a strained buffer....

About 10 shots of the quickest my D200 will do were done in RAW, compressed RAW and in large JPG.   I perceived no difference in speed of shots in this simple hands on test.  I am not sure when I would actually need to do this sort of photography.

As for compressed and full RAW, the most meaningful testing to me would be to lay an uncompressed over a compressed version and see if differences are apparent in the highlights or elsewhere.

Yes Tom on the Canon there is a way to shoot Raw and jpegs together you can then place them side by side and blow up to higher size and then view the results that way

LOl you guys should use Olympus no problems what so ever ;-)

Yes Tom on the Canon there is a way to shoot Raw and jpegs together
you can then place them side by side and blow up to higher size and
then view the results that way

I think the question is more about RAW vs. compressed RAW
rather than RAW vs. JPEG. 

--Martin

Does the Olympus handle this in a different way then Peter?

I would like to know that, too.  Curiosity beckons.

Apart from the fact that ORF are a lot more flexable, with all due respect, you guys are missing the point. Frame rate, buffer zones WTF???  I think you guys are losing touch with what photography is about.

Nikon v Canon, Canon v Nikon, buy an Olympus and you don't have to get involved in that stuff.
It's like buying a Harley or a Yamaha Virgo they both do the same things, they both even look similar but with one (or two) you have to go through all the crap and who says it's better? the pros who get a kick back and can claim tax on their investment??.
I know that Nikon and Canon have bigger lenses and Olympus are a visually smaller equivilent but isn't  it how you use the thing that counts??

The above is not a personnel attack on anybody, it's about the arrogance of the brands.

By the way you should not compress RAW in camera if you do then again you miss the point.
My humble advise is to, buy more cards and make them fast and think/ plan before you shoot ; simple :-)

ORF was a notifiable disease the last time I checked Who is getting into a debate about which is best here apart from you obvious Bias to Olympus The point of this thread was about if Raw was any better or not in the compressed format regardless of brands.. You made a sweeping statement that somehow Olympus was superior in some way?  I think at least we can all agree that it is not.. and leave it at that.  Cards being fast have little or no impact on how fast you can shoot regardless or camera make !!

Cards being fast have little or no impact on how fast you can
shoot regardless or camera make !!

My Kodak Z1015 IS camera can shot 720p HDTV at 30 fps provided
I use a 60X SDHC memory device (class 4). 

Device speed does matter. 

--Martin

for Movies I would think that is true Martin and for reading the card from the PC

d-SLR Timing Performance Parameter Nikon D70 Canon Digital Rebel Nikon D100 Canon
EOS-10D Startup "Instant" 3.09 0.63 2.32 AF Lag 0.34-0.49 0.25-0.28 0.150 0.146 Prefocus Lag 0.124 0.142 0.100 0.104 Cycle Time 0.34
(2.92 fps) 0.40
(2.50 fps) 0.35
(2.88 fps) 0.34
(2.94 fps) Buffer Depth 9 - 100+ 4 7-9 0
This is one test with the fastest cards in situ and the camera are rated at 3fps the end results show that the camera is not going to be faster than the quoted top cycle  speed. remains constant despite variations in lag times etc  

OK

Still camera photography ...

(1) For the data you supplied, are we talking about jpeg or RAW formatted images? 

(2) If jpeg, does the camera change the jpeg compression when a faster card is available?  It is possible that with a faster card, the camera will save larger files. 

(3) Camera autofocus time is another factor ... that would be constant no mater what the memory card speed ... in fact, that could dominate the performance numbers ... and might obscure the timing improvement granted by a faster memory device. 

--Martin

Yes I understand your logic Martin from what I  know it seems regardless of Jpeg or raw files reading to the camera buffer and speed of all reading and writing to the card may well be quicker in faster cards What I am saying is that it does not seem to make a difference to the camera performance in frames per  sec and cycle rates are unaffected the camera is rated at say 5 frames a sec thats all you will get regardless...

If you are saying that faster cards improve this I would like to know where this data is, as I have not found that the frame rate will improve which for me as a Bird in flight photographer being of special interest to me..  I have had to accept the limitations of my 5D at 3fps and I  would have to buy a 1Dmk11 or 111 if  I wanted to improve on this aspect..  I have no idea if this is another story for say Olympus or Nikon Owners or any other camera maker..

... frame rate will improve which for me as a Bird in flight
photographer being of special interest to me

The Kodak Z1015 IS camera is a

10MP still camera and also
30fps 720p HDTV camcorder with
15X optical zoom. 

It has 64 meg internal memory and accepts up to 16GBy SDHC fkash memory. 

That having been said, I must add that the camera is not getting the precise focus that I expect; there are playback issues with the 720p HDTV; playback on a PC requires a 3.0 GHz processr.  There are also some technical issues. 

What I'm saying is that there are cameras out there that can record lots of detail at high frame rate. 

--Martin

Of course, mechanical functions do affect repeat speed.

My concern with this thread was quality of image. That was essentially answered and came truly down to "it is something one must try and see", realizing a different scene may not have the dynamic range to be visibly affected by compression and another may be affected.  I have yet to prove that to myself or to try and do that.

With Danny doing in-flight photos, certainly rapid shooting coupled with rapid focus are most desired.   Even on servo (subject directed continuous focus),  it is easy to get an out of focus shot.  The speed limits seem here not so much memory card and buffer as mechanical limits.

As for buffer overload, I have pushed that shutter and clicked off about as many as I would likely need of a particular scene with a still camera, perhaps 10 or so.  With uncompressed or compressed there seems no real discernable difference in speed and no buffer overload.  Of course, I rarely shoot with rapid succession and some of you may do that with many more than 10 shots at  a session...

Yes, compressed or not, I do find composure the first aspect of a good shot, no one here is arguing that point.. I have not forgotten what photography is about...But with a flying bird sometimes it is simply keeping it in the focus area with continuous focus and rapid shots while wildly trying to stay on target   : )     Some photographers have shown ability to do that much better than the less practiced rest of us.        Tom.